TGC20ReId: A dataset for sport event re-identification in the wild

Peñate-Sánchez, Adrián; Freire-Obregón, David; Lorenzo-Melián, Adrián; Lorenzo-Navarro, Javier; Castrillón-Santana, Modesto

doi:10.1016/j.patrec.2020.08.003

Cited by 14 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Dataset To evaluate below the ultra-runners performance variation, we have partially used the dataset published by Penate et al [24]. The mentioned dataset was collected during an ultra-running competition, known as Transgrancanaria (TGC), held in March 2020.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Decontextualized I3D ConvNet for ultra-distance runners performance analysis at a glance

Freire-Obregón,

Lorenzo-Navarro,

Castrillón-Santana

2022

Preprint

Self Cite

View full text Add to dashboard Cite

In May 2021, the site runnersworld.com published that participation in ultra-distance races has increased by 1,676% in the last 23 years. Moreover, nearly 41% of those runners participate in more than one race per year. The development of wearable devices has undoubtedly contributed to motivating participants by providing performance measures in real-time. However, we believe there is room for improvement, particularly from the organizers point of view. This work aims to determine how the runners performance can be quantified and predicted by considering a non-invasive technique focusing on the ultra-running scenario. In this sense, participants are captured when they pass through a set of locations placed along the race track. Each footage is considered an input to an I3D ConvNet to extract the participant's running gait in our work. Furthermore, weather and illumination capture conditions or occlusions may affect these footages due to the race staff and other runners. To address this challenging task, we have tracked and codified the participant's running gait at some RPs and removed the context intending to ensure a runner-of-interest proper evaluation. The evaluation suggests that the features extracted by an I3D ConvNet provide enough information to estimate the participant's performance along the different race tracks.

show abstract

Section: Methodsmentioning

confidence: 99%

“…RPs locations, times and kilometer (extracted from[24]). In this work we have considered the locations in bold type.…”

mentioning

confidence: 99%

Decontextualized I3D ConvNet for ultra-distance runners performance analysis at a glance

Freire-Obregón,

Lorenzo-Navarro,

Castrillón-Santana

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…To the best of our knowledge, the most relevant public sports-focused re-id contribution that we found is the work in [17] that creates a dataset for re-id in long-distance running. Research on re-id in sports is very limited.…”

Section: Related Workmentioning

confidence: 99%

Sports Re-ID: Improving Re-Identification Of Players In Broadcast Videos Of Team Sports

Comandur¹

2022

Preprint

View full text Add to dashboard Cite

Re-identification (re-id) of people in images is a well-studied problem in computer vision for many applications. This work focuses on player re-identification in broadcast videos of team sports. Specifically, we focus on identifying the same player in images captured from different camera viewpoints during any given moment of a match. This task differs from traditional applications of person re-id in a few important ways. Firstly, players from the same team wear highly similar clothes, such as a team jersey/uniform, thereby making it harder to tell them apart. Secondly, there are only a few number of samples for each identity, which makes it harder to train a re-id system. Thirdly, the resolutions of the images are often quite low and vary a lot. This combined with heavy occlusions and fast movements of players greatly increase the challenges for re-id. In this paper, we propose a simple but effective hierarchical data sampling procedure and a centroid loss function that, when used together, increase the mean average precision (mAP) by 7 -11.5 and the rank-1 (R1) by 8.8 -14.9 without any change in the network or hyper-parameters used. Our data sampling procedure improves the similarity of the training and test distributions, and thereby aids in creating better estimates of the centroids of the embeddings (or feature vectors). Surprisingly, our study shows that in the presence of severely limited data, as is the case for our application, a simple centroid loss function based on euclidean distances significantly outperforms the popular tripletcentroid loss function. Our proposals provide comparable improvements for both convolutional networks and vision transformers. Our approach is currently ranked #1 on the ongoing SoccerNet Re-Identification Challenge 2022 leaderboard (testsplit) with a mAP of 86.0 and a R1 of 81.5. On the sequestered challenge split, we achieve an mAP of 84.9 and a R1 of 80.1. While we demonstrate results on soccer matches, our proposals naturally extend to any team sport. Research on re-id for sports-related applications is very limited and our work presents one of the first discussions in the literature on this. Code will be available in the supplementary material.Preprint. Under review.

show abstract

“…To overcome the possibility of bib number occlusion, some authors have combined face and bib number cues to improve overall performance [36]. Other authors have explored the body appearance to re-identify individuals in different locations, separated up to 20 hours, based on the body appearance [28]. Their results suggest a poor performance of standard re-identification techniques due to severe illumination changes, scenario changes, clothes changes, and even similar garments (people from the same club/team) Contrary to the previously described approach, the gait trait has also been adopted recently [8].…”

Section: Biometrics In Running Competitionsmentioning

confidence: 99%

“…The tracker is applied using an annotated bounding box, which comprises information about the source frame and the capture timestamp. The runner's identity is manually annotated based on our previous research works [13,28] after applying pedestrian detection at 1fps. Additional runner samples are computed by processing up to seven seconds around the annotated runner bounding box.…”

Section: Face Processing Pipelinementioning

confidence: 99%

Facial expression analysis in a wild sporting environment

Santana

Freire-Obregón

Hernández-Sosa

et al. 2022

Multimed Tools Appl

Self Cite

View full text Add to dashboard Cite

The scientific community and mass media have already reported the use of nonverbal behavior analysis in sports for athletes’ performance. Their conclusions stated that certain emotional expressions are linked to athlete’s performance, or even that psychological strategies serve to improve endurance performance. This paper examines the portrayal of well-known emotions and their relationship to the participants of an ultra-distance race in a high-stake environment. For this purpose, we analyzed almost 600 runners captured when they passed through a set of locations placed along the race track. We have observed a correlation between the runners’ facial expressions and their performance along the track. Moreover, we have analyzed Action Unit activations and aligned our findings with the state-of-the-art psychological baseline.

show abstract

TGC20ReId: A dataset for sport event re-identification in the wild

Cited by 14 publications

References 10 publications

Decontextualized I3D ConvNet for ultra-distance runners performance analysis at a glance

Decontextualized I3D ConvNet for ultra-distance runners performance analysis at a glance

Sports Re-ID: Improving Re-Identification Of Players In Broadcast Videos Of Team Sports

Facial expression analysis in a wild sporting environment

Contact Info

Product

Resources

About